Title: Anomaly Detection for Autonomous Driving
Speaker: Abdullah Al Redwan Newaz, Ph.D.
Date: Friday, September 25, 2020
SYNOPSIS
In the realm of signal processing for driving behavior analysis, anomaly detection is an efficient method for identifying the inconsistent behaviors of the drivers. It can allow to determine if an abnormal driving behavior led to a road accident. Therefore, efficient anomaly identification in driving behaviors is one of the crucial steps for ensuring safe driving. In particular, the advent of autonomous driving technologies would lead to a transportation system with mixed manned and unmanned traffic vehicles, for which automatically identifying the common anomalies in human driving behavior will be a crucial safety requirement.
Anomaly detection has been researched in the literature from different perspectives. In this research, however, we are interested in driving behavior which can be defined as multi-dimensional time series data that has been recorded by using various types of sensors. Since the information from each sensor type constitutes a dimension of the driving behavior data, it is important to extract the necessary and sufficient features to model the driving behavior. Therefore, the problem of driving behavior modeling is how to extract the essential features using sensor data from different modalities. On the other hand, to detect the anomalies, we utilize deep neural networks that can learn a normal driving behavior distribution during the training phase, and then, identify the anomalies in the testing phase.
ABOUT THE SPEAKER
Abdullah Al Redwan Newaz is a Postdoctoral Research Associate at Autonomous Cooperative Control of Emergent Systems of Systems (ACCESS) Laboratory at North Carolina A&T State University. Before that, he was a Postdoctoral Researcher at Rice University, Houston, TX, USA, and Nagoya University, Nagoya, Japan, in 2018-2020, 2017-2018, respectively. He earned Ph.D. and M.S. degrees in Information Science from Japan Advanced Institute of Science and Technology, Ishikawa, Japan, in 2017 and 2014, respectively. He received a B.Sc. degree in Mechanical Engineering from Rajshahi University of Engineering and Technology, Rajshahi, Bangladesh, in 2011. His research interests include autonomous systems, applied machine learning, motion planning under uncertainty, optimal control, policy synthesis, model checking, and related domains. He is an IEEE professional member and serves as reviewer for top-tier conferences and journals.